So, I've cleaned the code a little, using some suggestions
given earlier by Rune and Mark.

Particle start with a smooth-turbulenced initial direction,
but tend to move with the actual wind-direction (<1,0,0>)
after a given amount of time.

Result:
Particles spread out, at first smoothly, but it looks chaotic in
the end.
You recognize where particles begin to move with the wind,
as the expansion stops.

All in all, it just looks crappy. The algorithm isn't (in this system)
worth the effort. I could just randomly turbulence, or add
some random vector to be added to the position in the first
few seconds of a particle's lifetime, and it would
look the same. :-(

So I've decided to work on it so that it may be used to
create varying wind-direction and speed over time.
Sadly, non-I/O binds me to the issue that particles
may start with different initial settings, but when
you see a particle blown with wind-speeds of
5mph and a particle passing it with higher speed,
it would just look stupid.

So, finally, I decided to discard that idea. I'm pondering
a little about a different approach, but right now, I think
I'll just leave it.
It gave me some more ideas for an I/O system, as well
as some other nice stuff (for example leaves falling
to the ground tumbling), but we'll see.

Thanks for comments anyway!

--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html

Post a reply to this message

Attachments:
Download 'smoke2.mpg' (193 KB)

Tim Nikias wrote:

> You recognize where particles begin to move with the wind,
> as the expansion stops.

This does look a bit odd.

> All in all, it just looks crappy. The algorithm isn't (in this system)
> worth the effort. I could just randomly turbulence, or add
> some random vector to be added to the position in the first
> few seconds of a particle's lifetime, and it would
> look the same. :-(

It might not be the best smoke effecct, but I still think that if you put a good
texture on those spheres that made it _look_ like smoke (esp. fading with time), and
shot them out the back of the rocket (not stationary as you have done here), alot of
the problems would be less noticeable. As with most computer generated graphics, it
doesn't have to be a rigorous simulation of smoke, it just has to _look_ like smoke.

> So I've decided to work on it so that it may be used to
> create varying wind-direction and speed over time.
> Sadly, non-I/O binds me to the issue that particles
> may start with different initial settings, but when
> you see a particle blown with wind-speeds of
> 5mph and a particle passing it with higher speed,
> it would just look stupid.

Why would it look stupid? This happens in real life sometimes.

Take a look at my reply to your first smoke anim too. I have posted a v. short anim of
my bozo technique you might want to look at.

MJL


--
text{ttf"timrom.ttf"concat(#local O=1;#while(O<7)chr(val(substr(concat(#local Q=
1;#while(Q<7)str(asc(substr("???<?>",Q,1))-56,0,0),#local Q=Q+1;#end""),O,2))),#
local O=O+2;#end"").1,0pigment{rgb 9}translate-<1,.3,-2>} // MJL

Post a reply to this message

Well, okay, I won't drop it. I'll work on it a little
more and post a fancy animation the next time, and
perhaps then everything will be fine. At least it
will _look_ fine.

Thanks.


--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html

Post a reply to this message